POD-assisted strategies for structural topology optimization

We propose a new numerical tool for structural optimization design. To cut down the computational burden typical of the Solid Isotropic Material with Penalization (SIMP) method, we apply Proper Orthogonal Decomposition on SIMP snapshots computed on a fixed grid to construct a rough structure (predictor) which becomes the input of a SIMP procedure performed on an anisotropic adapted mesh (corrector). The benefit of the proposed design tool is to deliver smooth and sharp layouts which require a contained computational effort before moving to the 3D printing production phase.     

不同粒度纳米碳酸钙的可控制备

不同粒度的纳米碳酸钙具有不同的特性和应用范围,目前不同粒度纳米碳酸钙的可控制备方法还未见报道。以氯化钙、碳酸铵为原料,以乙醇、柠檬酸和焦磷酸钠为分散剂,采用复分解法研究了不同制备条件对纳米碳酸钙粒径的影响规律,在此基础上制备出平均粒径为17~71 nm的碳酸钙,并对其进行了表征。实验结果表明,所制备的纳米碳酸钙均为球霰石型,纯度很高,而且形貌近似球形。制备条件对纳米碳酸钙的粒径有显著的影响：随着反应温度升高,纳米碳酸钙的粒径减小;随着氯化钙滴加时间的增加,纳米碳酸钙的粒径增大;随着反应物浓度的提高,纳米碳酸钙的粒径先减小后增大;当分散剂为柠檬酸和焦磷酸钠的混合溶液时所得纳米碳酸钙的粒径最小,当分散剂为乙醇和焦磷酸钠的混合溶液时所得纳米碳酸钙的粒径最大。利用这些影响规律通过控制制备条件可以实现所需粒径纳米碳酸钙的可控制备。     

Mechanism of methanol decomposition on the Pd/WC(0001) surface unveiled by first-principles calculations

In this study, the decomposition of methanol into the CO and H species on the Pd/tungsten carbide (WC)(0001) surface is systematically investigated using periodic density functional theory (DFT) calculations. The possible reaction pathways and intermediates are determined. The results reveal that saturated molecules, i.e., methanol and formaldehyde, adsorb weakly on the Pd/ WC(0001) surface. Both CO and H prefer three-fold sites, with adsorption energies of −1.51 and −2.67 eV, respectively. On the other hand, CH₃O stably binds at three-fold and bridge sites, with an adsorption energy of −2.58 eV. However, most of the other intermediates tend to adsorb to the surface with the carbon and oxygen atoms in their sp³ and hydroxyl-like configurations, respectively. Hence, the C atom of CH₂OH preferentially attaches to the top sites, CHOH and CH₂O adsorb at the bridge sites, while COH and CHO occupy the three-fold sites. The DFT calculations indicate that the rupture of the initial C–H bond promotes the decomposition of CH₃OH and CH₂OH, whereas in the case of CHOH, O–H bond scission is favored over the C–H bond rupture. Thus, the most probable methanol decomposition pathway on the Pd/WC(0001) surface is CH₃OH → CH₂OH → trans-CHOH → CHO → CO. The present study demonstrates that the synergistic effect of WC (as carrier) and Pd (as catalyst) alters the CH₃OH decomposition pathway and reduces the noble metal utilization.     

鲸鱼优化支持向量机的短期风电功率预测

为提高风电预测的精度,提出一种鲸鱼优化支持向量机SVM(support vector machine)的组合预测模型。该模型针对风电序列的非平稳波动特性,首先应用集合经验模态分解技术EEMD(ensemble empirical mode de?composition)将原始风电序列分解为一系列不同特征尺度的子序列;并引入鲸鱼优化算法WOA(whales optimiza?tion algorithm)解决SVM中学习参数选择难的问题,进而对各子序列建立WOA_SVM预测模型;最后,叠加各子序列的预测值以得到最终预测值。仿真表明,所提EEMD_WOA_SVM模型具有较高的风电预测精度,显著优于其他基本模型。     

多共振分量融合CNN的行星齿轮箱故障诊断

针对行星齿轮箱中各部件所激起的振动成分混叠、早期故障特征经常被较强的各级齿轮谐波成分以及环境噪声所湮没的问题，提出一种多共振分量融合卷积神经网络（multi-resonance component fusion based convolutional neural network，简称MRCF-CNN）的行星齿轮箱故障诊断方法。首先，对振动信号进行共振稀疏分解，得到包含齿轮谐波成分的高共振分量和可能包含轴承故障冲击成分的低共振分量；其次，构建多共振分量融合卷积神经网络，将得到的高、低共振分量和原始振动信号进行自适应的特征级融合，通过有监督的方式训练模型并进行行星齿轮箱故障诊断。对行星齿轮箱实验数据的分析结果表明，该方法能够有效分类行星齿轮箱中滚动轴承和齿轮的故障，成功对行星齿轮箱故障进行诊断，同时能够进一步增强卷积神经网络对振动信号所蕴含的故障信息的辨识能力。     

Concept design of a novel reformer producing hydrogen for internal combustion engines using fuel decomposition method: Performance evaluation of coated monolith suitable for on-board applications

Hydrogen addition effectively reduces the fuel consumption of spark ignition engines. We propose a new on-board reformer that produces hydrogen at high concentrations and enables multi-mode operations. For the proposed reformer, we employ a catalytic fuel decomposition reaction via a commercial NiO–CaAl₂O₄ catalyst. We explore the physical and chemical aspects of the reforming process using a fixed bed micro-reactor operating at temperatures of 550–700 °C. During reduction, methane is decomposed to form hydrogen and carbon. Carbon formation is critical to hydrogen production, and free space for carbon growth is essential at low temperatures (≤600 °C). We define a new accumulated conversion ratio that quantitatively measures highly transient catalytic decomposition. The free space of the coated monolith clearly aided low-temperature decomposition with negligible pressure drop. The coated substrate is therefore suitable for on-board applications considering that our reformer concept also utilizes the catalytic fuel decomposition reaction.     

Cobalt doping of α-Fe/Al2O3 catalysts for the production of hydrogen and high-quality carbon nanotubes by thermal decomposition of methane

Fe–Co/Al₂O₃ catalysts were developed and tested in the catalytic decomposition of methane (CDM) for the synthesis of multi-wall carbon nanotubes (MWCNT) and the CO₂-free hydrogen production. While Fe (54.5–66.7 mol.%) is the main active phase for the carbon formation on the catalyst, Co acts as dopant aiming to improve its overall catalytic behaviour. Catalysts with Co contents of up to 18.2 M% showed the presence of α-Fe and Fe–Co crystallites with different size and lattice parameter. Fe_1-xCo_x alloy with bcc crystal system was identified only for Co contents of 14.0% and above, and presented a lattice constant lower than α-Fe, which would modify the carbon diffusion of the metal particle during the MWCNT growth. Co inhibited the Fe₃C formation during CDM resulting in higher carbon formations and longer activity times. This phase, shown in undoped catalysts, favored the presence of bamboo-type carbon nanotubes.     

Estimating posterior inference quality of the relational infinite latent feature model for overlapping community detection

Overlapping community detection has become a very hot research topic in recent decades, and a plethora of methods have been proposed. But, a common challenge in many existing overlapping community detection approaches is that the number of communities K must be predefinedmanually. We propose a flexible nonparametric Bayesian generative model for count-value networks, which can allow K to increase as more and more data are encountered instead of to be fixed in advance. The Indian buffet process was used to model the community assignment matrix Z, and an uncollapsed Gibbs sampler has been derived.However, as the community assignment matrix Z is a structured multi-variable parameter, how to summarize the posterior inference results and estimate the inference quality about Z, is still a considerable challenge in the literature. In this paper, a graph convolutional neural network based graph classifier was utilized to help to summarize the results and to estimate the inference quality about Z. We conduct extensive experiments on synthetic data and real data, and find that empirically, the traditional posterior summarization strategy is reliable.     

基于改进ReaxFF 力场的全氟己酮热分解机理研究

采用基于密度泛函的紧束缚模型（DFTB）和改进的反作用力场（ReaxFF）对全氟己酮的热分解机理进行反应分子动力学仿真研究，分析了热分解的初始反应路径、微观反应过程以及体系内气体的生成机理，探讨了温度对全氟己酮热分解的影响，建立了以仿真数据为基础的一级动力学方程并对热分解过程进行一级动力学评估。结果表明，全氟己酮热分解的起始反应为：全氟己酮→全氟异戊烷+一氧化碳，全氟己酮热分解的主要产物为一氧化碳、四氟甲烷、六氟丙烯、全氟异丁烯、十氟丁烷、全氟异戊烷及大量活性基团。经一级动力学评估得到指前因子和表观活化能分别为5.021 7×1014 s-1和1 771.96 kJ/mol。     

基于VMD瞬时能量与GA- RBF的滚动轴承故障诊断

针对滚动轴承早期故障的有效识别,提出了一种基于VMD瞬时能量与GA优化的RBF神经网络的滚动轴承故障诊断方法,可以有效对滚动故障做出诊断。首先,VMD将滚动轴承振动信号进行分解成合适数目的本证模态函数;其次,计算本证模态函数分量的瞬时能量并组成特征向量;最后,将特征向量输入到GA优化的RBF神经网络实现轴承故障识别。通过滚动轴承故障诊断实验对该方法进行验证。结果表明,该方法识别滚动轴承故障的准确率为96.43%,较默认参数的RBF神经网络和EEMD瞬时能量与GA-RBF神经网络有明显的提高,证明了所提方法的可行性。